From time-series to 2D images for building occupancy prediction using deep transfer learning

<p dir="ltr">Building occupancy information could aid energy preservation while simultaneously maintaining the end-user comfort level. Energy conservation becomes essential since energy resources are scarce and human dependency on appliances is only exponentially increasing. While in...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Aya Nabil Sayed (17317006) (author)
مؤلفون آخرون: Yassine Himeur (14158821) (author), Faycal Bensaali (12427401) (author)
منشور في: 2023
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513543611613184
author Aya Nabil Sayed (17317006)
author2 Yassine Himeur (14158821)
Faycal Bensaali (12427401)
author2_role author
author
author_facet Aya Nabil Sayed (17317006)
Yassine Himeur (14158821)
Faycal Bensaali (12427401)
author_role author
dc.creator.none.fl_str_mv Aya Nabil Sayed (17317006)
Yassine Himeur (14158821)
Faycal Bensaali (12427401)
dc.date.none.fl_str_mv 2023-03-01T18:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.engappai.2022.105786
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/From_time-series_to_2D_images_for_building_occupancy_prediction_using_deep_transfer_learning/24474652
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Engineering
Electronics, sensors and digital hardware
Environmental engineering
Information and computing sciences
Artificial intelligence
Data management and data science
Machine learning
Occupancy detection
Environmental data
Feature engineering
Image transformation
Deep learning
Convolutional neural network
dc.title.none.fl_str_mv From time-series to 2D images for building occupancy prediction using deep transfer learning
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Building occupancy information could aid energy preservation while simultaneously maintaining the end-user comfort level. Energy conservation becomes essential since energy resources are scarce and human dependency on appliances is only exponentially increasing. While intrusive sensors (i.e., cameras and microphones) can raise privacy concerns, this paper presents an innovative non-intrusive occupancy detection approach using environmental sensor data (e.g., temperature, humidity, carbon dioxide (CO<sub>2</sub>), and light sensors). The proposed scheme transforms multivariate time-series data into images for better encoding and extracting relevant features. The utilized image transformation method is based on data normalization and matrix conversion. Precisely, by representing time-series in 2D space, an encoding kernel can move in two directions while it can move only in one direction when applied to a 1D signal. Moreover, machine learning (ML) and deep learning (DL) techniques are utilized to classify occupancy patterns. Several simulations are used to evaluate the approach; mainly, we investigated pre-trained and custom convolutional neural network (CNN) models. The latter attained an accuracy of 99.00%. Additionally, pixel data are extracted from the generated images and subjected to traditional ML methods. Throughout the numerous comparison settings, it was observed that the latter strategy provided the optimal balance of 99.42% accuracy performance and minimal training time across the occupancy datasets.</p><h2>Other Information</h2><p dir="ltr">Published in: Engineering Applications of Artificial Intelligence<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.engappai.2022.105786" target="_blank">https://dx.doi.org/10.1016/j.engappai.2022.105786</a></p>
eu_rights_str_mv openAccess
id Manara2_0639dd4b4eeb90ab691041363b17c1ca
identifier_str_mv 10.1016/j.engappai.2022.105786
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/24474652
publishDate 2023
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling From time-series to 2D images for building occupancy prediction using deep transfer learningAya Nabil Sayed (17317006)Yassine Himeur (14158821)Faycal Bensaali (12427401)EngineeringElectronics, sensors and digital hardwareEnvironmental engineeringInformation and computing sciencesArtificial intelligenceData management and data scienceMachine learningOccupancy detectionEnvironmental dataFeature engineeringImage transformationDeep learningConvolutional neural network<p dir="ltr">Building occupancy information could aid energy preservation while simultaneously maintaining the end-user comfort level. Energy conservation becomes essential since energy resources are scarce and human dependency on appliances is only exponentially increasing. While intrusive sensors (i.e., cameras and microphones) can raise privacy concerns, this paper presents an innovative non-intrusive occupancy detection approach using environmental sensor data (e.g., temperature, humidity, carbon dioxide (CO<sub>2</sub>), and light sensors). The proposed scheme transforms multivariate time-series data into images for better encoding and extracting relevant features. The utilized image transformation method is based on data normalization and matrix conversion. Precisely, by representing time-series in 2D space, an encoding kernel can move in two directions while it can move only in one direction when applied to a 1D signal. Moreover, machine learning (ML) and deep learning (DL) techniques are utilized to classify occupancy patterns. Several simulations are used to evaluate the approach; mainly, we investigated pre-trained and custom convolutional neural network (CNN) models. The latter attained an accuracy of 99.00%. Additionally, pixel data are extracted from the generated images and subjected to traditional ML methods. Throughout the numerous comparison settings, it was observed that the latter strategy provided the optimal balance of 99.42% accuracy performance and minimal training time across the occupancy datasets.</p><h2>Other Information</h2><p dir="ltr">Published in: Engineering Applications of Artificial Intelligence<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.engappai.2022.105786" target="_blank">https://dx.doi.org/10.1016/j.engappai.2022.105786</a></p>2023-03-01T18:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.engappai.2022.105786https://figshare.com/articles/journal_contribution/From_time-series_to_2D_images_for_building_occupancy_prediction_using_deep_transfer_learning/24474652CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/244746522023-03-01T18:00:00Z
spellingShingle From time-series to 2D images for building occupancy prediction using deep transfer learning
Aya Nabil Sayed (17317006)
Engineering
Electronics, sensors and digital hardware
Environmental engineering
Information and computing sciences
Artificial intelligence
Data management and data science
Machine learning
Occupancy detection
Environmental data
Feature engineering
Image transformation
Deep learning
Convolutional neural network
status_str publishedVersion
title From time-series to 2D images for building occupancy prediction using deep transfer learning
title_full From time-series to 2D images for building occupancy prediction using deep transfer learning
title_fullStr From time-series to 2D images for building occupancy prediction using deep transfer learning
title_full_unstemmed From time-series to 2D images for building occupancy prediction using deep transfer learning
title_short From time-series to 2D images for building occupancy prediction using deep transfer learning
title_sort From time-series to 2D images for building occupancy prediction using deep transfer learning
topic Engineering
Electronics, sensors and digital hardware
Environmental engineering
Information and computing sciences
Artificial intelligence
Data management and data science
Machine learning
Occupancy detection
Environmental data
Feature engineering
Image transformation
Deep learning
Convolutional neural network